Quote from: BurtW on August 03, 2014, 05:47:19 PM

Quote from: X7 on August 03, 2014, 05:00:01 PM

It is silly to assume that by such a date we would not have much more efficient ways of powering this technology with the vast amount of money which IS being and WILL be poured into the infrastructure of maintaining this epic ecosystem.

2/10 for effort

Quote from: Anders on August 03, 2014, 05:27:03 PM

I hope those who did the math included this:

"Combined with Moore's law, performance per watt would grow at roughly the same rate as transistor density, doubling every 12 years." -- http://en.wikipedia.org/wiki/Moore%27s_law#Other_formulations_and_similar_laws

Neither one of you saw or understood that mining efficiency does not matter when calculating the total power consumption of the Bitcoin Network.

Efficiency only affects total hash rate and difficulty, not power consumption.

It is a given that mining hardware will become more efficient for the reasons you stated. However, more efficient mining hardware means that for a given amount of power the network will just produce a larger hash rate. When more efficient mining hardware hits the market miners will just buy more of it until they consume the same amount of power as before the more efficent mining hardware was on the market.

The total power consumed by the network is only bounded by the amount of money available in each block to spend on power.

Couple of things:

1) That assumes an even distribution of "most efficient" mining hardware, and that access to available power is uniformly distributed. Right? I don't think that's likely - and "When more efficient mining hardware hits the market miners will just buy more of it until they consume the same amount of power as before" is assuming that anyone can get their hands on the capital/BTC to keep upgrading, and that they'll never be told "nope, no more power for you" from the electric company. It's more likely that people who would be otherwise be spending lots on power will be able afford very efficient mining hardware that doesn't use much power, and the rest will use inefficient hardware and soak up as much power as they can get. I think this makes it hard to predict how much power consumption is for a certain target hash rate. At least I think that's true... check my logic please?

2) Morgan Stanley's just released a report re: Tesla and home solar, suggesting that available power will increase by 39GW a year through 2020. http://forms.greentechmedia.com/Extranet/95679/Morgan%20Stanley%20Solar%20Power%20&%20Energy%20Storage%20Blue%20Paper%20July%2029%202014.pdf So new power is coming online all the time apparently

(Though if I'm right, then this means Russia / Saudi / China / some other major energy producer and user will have to a) replace their dependence on using and selling existing fuels with new technologies and b) shift a ton of their new surplus power straight to mining BTC... like, mining would have to happen on the nation-state or Google/Apple corporate level. We're not getting this much power from nerds with asics anymore, it's gotta be datacenters in remote cold locations where energy can be spent solely on mining and not on cooling...)

If I were Putin I'd open a colo in Siberia and send 10% of Gazprom's reserves straight to it to power my ASIC farm /conspiracy Wink